Qwen3.8-27B Released with Native Multimodal Support and 262K Context
const_reborn · x · 2026-08-15
Alibaba has released the Qwen3.8-27B model with open weights. It is a native multimodal dense model with 27B parameters, outperforming Qwen3.7-Plus overall, particularly in real-world coding and office workflows. It features 262K native context, extendable to 1M tokens via YaRN, and is licensed under Apache 2.0. The open weights for Qwen3.8-2.4T-A95B have also been released. engy.ai has launched the inference API for Qwen3.8-27B at $0.045/M input and $0.32/M output.
Related event: Alibaba Open-Sources Qwen3.8-27B Multimodal Model That Runs on a Single GPU(2 posts)→
More from Models
- User praises LLM excessively to prevent apologetic regression mode — wavefnx · 2026-08-15
- User Feedback: Qwen3.8 Overthinks and Misses the Point on Tasks — WhatererBlah555 · 2026-08-15
- Qwen3.8 27B评测：一次性编程能力显著提升 — kms_dev · 2026-08-15
- Whisper: Local speech-to-text system for multiple languages — ZabihullahAtal · 2026-08-15
- Fable is top but token-heavy; GPT models more efficient with similar intelligence — gethackteam · 2026-08-15
- Qwen3.8-35B-A3B Spotted in GitHub Code, Optimized for Consumer GPUs — cephaloform · 2026-08-15